Upload
Login or register

HypoLast

Last status update:
-
Gender: male
Age: 34
Date Signed Up:9/29/2009
Last Login:7/29/2016
FunnyJunk Career Stats
Comment Ranking:#13037
Highest Content Rank:#6561
Highest Comment Rank:#13038
Content Thumbs: 41 total,  83 ,  42
Comment Thumbs: 283 total,  349 ,  66
Content Level Progress: 74.57% (44/59)
Level 0 Content: Untouched account → Level 1 Content: New Here
Comment Level Progress: 20% (2/10)
Level 127 Comments: Respected Member Of Famiry → Level 128 Comments: Respected Member Of Famiry
Subscribers:0
Content Views:5822
Times Content Favorited:1 times
Total Comments Made:118
FJ Points:331

  • Views: 1980
    Thumbs Up 12 Thumbs Down 4 Total: +8
    Comments: 11
    Favorites: 1
    Uploaded: 12/18/15
    No admins allowed! No admins allowed!
  • Views: 521
    Thumbs Up 24 Thumbs Down 20 Total: +4
    Comments: 7
    Favorites: 0
    Uploaded: 12/06/11
    Chair Chair

latest user's comments

#17 - yeah it's funny 06/05/2016 on Boom! HEADSHOT! +8
#16 - yeah it's funny 06/05/2016 on Boom! HEADSHOT! +13
#9 - Number 2 is magic the gathering 05/19/2016 on Dank D&D character concept... +1
#27 - Here's a video you may find interesting. Relevant to your question.  [+] (1 new reply) 02/19/2016 on Evolution 0
User avatar
#28 - bashmuffins (02/19/2016) [-]
I love how he said that Humans and Chimps are roughly the same as Horses and Zebras

But yes thank you for that very informative video.
#10 - If I go through the trouble of even making it a Chrome extensi… 12/18/2015 on No admins allowed! 0
#5 - Yes I am. Also this has already been changed. It chan…  [+] (3 new replies) 12/18/2015 on No admins allowed! -1
User avatar
#8 - tsagu (12/18/2015) [-]
Could you perhaps also make it for Firefox? I'd like to NOT use chrome since it's a resource hog.
User avatar
#10 - HypoLast (12/18/2015) [-]
If I go through the trouble of even making it a Chrome extension which I probably won't I doubt I would also go through the trouble of porting it to FireFox, sorry.
#9 - anon (12/18/2015) [-]
Chrome uses memory, Firefox uses CPU, IE just loads all scripts without optimization, thus slower. Edge is decent, but doesn't load certain types of code/webpages and addons are slim.

At the end of the day, every browser will use resources, Chrome seems to be comfortable merely because I have used it for time. The add-ons are rich and the content is quick and easy, the google login feature makes for easy syncing across computers -- yes I know you can do this on other browsers too.
#36 - Comment deleted 09/10/2015 on description please 0
#88 - It's been a while since I've checked Funnyjunk, so sorry for t…  [+] (1 new reply) 07/25/2015 on History of flash 0
#89 - gasur (07/28/2015) [-]
I see, thanks for teaching me
#85 - Web sockets are just TCP with a really short keep alive, which…  [+] (3 new replies) 07/21/2015 on History of flash 0
#86 - gasur (07/21/2015) [-]
What I'm worried about, isn't the extra bandwidth caused by overhead etc, but more of the packet guaranteed plus guaranteed packet order. That means it will wait for the correct packet in the sequence, which would make the video buffer. It happens not too often for me on Twitch, but I could imagine that for some it would be a nightmare.
User avatar
#88 - HypoLast (07/25/2015) [-]
It's been a while since I've checked Funnyjunk, so sorry for taking so long. You're right that the main concern is when the packets get held up by other ones. Basically what the low keep alive time on Websockets is doing is making it so that rather than waiting a few seconds for an ack signal (the client saying it received the information) it only waits a few hundred milliseconds, and redundancy is built in to future packets to make that time even shorter in many cases. That means that even in worst case scenarios where up to 5%+ of your packets are getting corrupted you'll rarely see more than a 200% increase in transfer time (keep in mind this is usually around 80ms in the first place, so pretty irrelevant) on small packets.

Any latency beyond that is a problem with bandwidth, which is a limitation regardless of your transfer protocol.
#89 - gasur (07/28/2015) [-]
I see, thanks for teaching me
#83 - Perhaps there's limitations to real time streaming that I don'…  [+] (5 new replies) 07/20/2015 on History of flash 0
#84 - gasur (07/20/2015) [-]
I would assume that Web Sockets etc, can only support TCP (due to the nature of the HTTP/TCP protocol). But could there be a way to circumvent that, without needing additional software such as Flash? As TCP isn't that great for streaming.
User avatar
#85 - HypoLast (07/21/2015) [-]
Web sockets are just TCP with a really short keep alive, which causes them to act like UDP in a lot of cases (though packet order is still guaranteed). Honestly since the internet is so fast now there isn't really that much difference unless you're doing time critical applications like games, but even then web sockets have proven that they're fast enough. Most people didn't notice at all when Youtube switched to HTML5, which is a testament to its streaming capabilities, so I don't think it's fair to say that TCP isn't good for streaming anymore

Like I mentioned WebRTC is trying to get UDP in HTML5 but there are security concerns that are holding it up (the potential for DDoS attacks, mostly), but a lot of people don't feel it's really necessary because of the options already available.
#86 - gasur (07/21/2015) [-]
What I'm worried about, isn't the extra bandwidth caused by overhead etc, but more of the packet guaranteed plus guaranteed packet order. That means it will wait for the correct packet in the sequence, which would make the video buffer. It happens not too often for me on Twitch, but I could imagine that for some it would be a nightmare.
User avatar
#88 - HypoLast (07/25/2015) [-]
It's been a while since I've checked Funnyjunk, so sorry for taking so long. You're right that the main concern is when the packets get held up by other ones. Basically what the low keep alive time on Websockets is doing is making it so that rather than waiting a few seconds for an ack signal (the client saying it received the information) it only waits a few hundred milliseconds, and redundancy is built in to future packets to make that time even shorter in many cases. That means that even in worst case scenarios where up to 5%+ of your packets are getting corrupted you'll rarely see more than a 200% increase in transfer time (keep in mind this is usually around 80ms in the first place, so pretty irrelevant) on small packets.

Any latency beyond that is a problem with bandwidth, which is a limitation regardless of your transfer protocol.
#89 - gasur (07/28/2015) [-]
I see, thanks for teaching me
[ 114 Total ]

user's friends